The housing demand analysis and prediction of the real estate based on the AWGM (1, N) model

Abstract

Purpose

The purpose of this paper is to seek to drive the modernization of the entire national economy and maintain in the long-term stability of the whole society; this paper proposes an improved model based on the first-order multivariable grey model [GM (1, N) model] for predicting the housing demand and solving the housing demand problem.

Design/methodology/approach

This paper proposes an improved model based on the first-order multivariable grey model [GM (1, N) model] for predicting the housing demand and solving the housing demand problem. First, a novel variable SW evaluation algorithm is proposed based on the sensitivity analysis, and then the grey relational analysis (GRA) algorithm is utilized to select influencing factors of the commodity housing market. Finally, the AWGM (1, N) model is established to predict the housing demand.

Findings

This paper selects seven factors to predict the housing demand and find out the order of grey relational ranked from large to small: the completed area of the commodity housing> the per capita housing area> the one-year lending rate> the nonagricultural population > GDP > average price of the commodity housing > per capita disposable income.

Practical implications

The model constructed in the paper can be effectively applied to the analysis and prediction of Chinese real estate market scientifically and reasonably.

Originality/value

The factors of the commodity housing market in Wuhan are considered as an example to analyze the sales area of the commodity housing from 2015 to 2017 and predict its trend from 2018 to 2019. The comparison between demand for the commodity housing actual value and one for model predicted value is capability to verify the effectiveness of the authors’ proposed algorithm.

Citation

Xiong, X., Guo, H. and Hu, X. (2021), "The housing demand analysis and prediction of the real estate based on the AWGM (1, N) model", Grey Systems: Theory and Application, Vol. 11 No. 2, pp. 222-240. https://doi.org/10.1108/GS-09-2019-0035